The principal goal of the original research plan was to search for reliable "markers" or indicators of vulnerability for alcoholism. Families were selected because some adult siblings were alcoholic and others were not, allowing us to trace the "marker" through families to determine its association with affected status. The unaffected siblings from these families who were selected through an alcoholic proband were also contrasted with siblings from control families (no first degree relatives with alcoholism). Our research during years 01-03 has suggested a number of biological indicators deserving of further study. One of these is measurement of the late components of the event-related brain potentials (e.g., P300), another is increased haplotype sharing (HLA) among concordant affected sibs in contrast to discordant sib pairs (all family members had well characterized diagnoses, e.g., Feighner Criteria). During the proposed renewal period we plan to ascertain the complete sibship, which with their spouses and children will provide us with a large number of nuclear families with which to perform segregation analysis. This analysis along with quantitative estimates of risk (10 year and lifetime) to these children (most have not entered the age of risk) for alcoholism will be determined. With this information, we can test the reliability of our putative indicators by performing correlational analyses between the estimate risk and our quantitative indicators. With follow-up during subsequent investigations, we can confirm whether or not our predictions for specific children are correct. Also, we will begin to explore the nature of the neurobehavioral deficits reported in the literature for children of alcoholics utilizing measures of vigilance, iconic memory, and neurological functioning including static ataxia.